LaunchKit · 2026
Back to Skills

Activate when the user needs to interact with any website — browser automation, web scraping, screenshots, form.

0
1.3k downloads
by @adcentury

Setup & Installation

openclaw skills install @adcentury/actionbook

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install actionbook

Version History

v0.1.1Feb 13, 2026 - Version note by skill developer:

Optimize the use cases and fallback handling

What This Skill Does

Actionbook is a library of pre-verified page interaction data for browser automation. It provides CLI commands to search for and retrieve page structures with working CSS selectors, then drive a browser to complete tasks on any website.

Pre-verified selectors and health scores remove the need to inspect page DOM manually, so browser automation scripts run without a separate discovery step.

When to use it

  • Automating job applications across multiple job boards
  • Scraping product prices from e-commerce sites on a schedule
  • Filling and submitting repetitive data entry forms
  • Taking scheduled screenshots for visual monitoring
  • Building end-to-end tests for external websites

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: Search arXiv for papers about Neural Networks in titles only

AGENT
  1. 1Runs `actionbook search` with a full-intent query scoped to arxiv.org
  2. 2Runs `actionbook get` on the result ID to retrieve page structure and CSS selectors
  3. 3Opens the browser and navigates to the arXiv advanced search URL
  4. 4Fills the search field and sets the title-only filter using selectors from the response
  5. 5Clicks submit, waits for navigation, and extracts page text
OUTPUT

List of matching arXiv paper titles and abstracts returned as page text